Explainable Lymph Node Diagnosis Model Combining MobileNet and ConvNextTiny

Amira Bouamrane; Makhlouf Derdour; Saad Harous; Kouzo Abdellah; Mohamed Deriche; Moustafa Sadek Kahil

doi:10.1109/PAIS66004.2025.11126497

Explainable Lymph Node Diagnosis Model Combining MobileNet and ConvNextTiny

Amira Bouamrane
, Makhlouf Derdour
, Saad Harous
, Kouzo Abdellah
, Mohamed Deriche
, Moustafa Sadek Kahil

College of Engineering and Information Technology

University of Oum El Bouaghi
University of Sharjah
University of Djelfa

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Computer-Aided Diagnostic (CADx) systems have proven effective in classifying pulmonary nodules. However, these models' reliability issue represents a topic of research and discussion. This paper proposes an interpretable lightweight hybrid model for diagnosing lymph nodes using computed tomography images and deep learning. The hybrid model combines MobileNetV3Small with ConvNextTiny to highlight features and improve performance by combining two datasets to assess data diversity, including LIDC-IDRI and the chest CT scan images for the lung cancer dataset. The model is evaluated using an external dataset to show its generalization capability. LIME is used to explain the model's decisions. The approach achieved a low false positive rate (FPR) and false negative rate (FNR) of 1.04%, resulting in high performance in all the metrics with an accuracy of 98.56% and a receiver operating characteristic (ROC) of 100% in the IQ-OTH/NCCD dataset.

Original language	English
Title of host publication	PAIS 2025 - Proceeding
Subtitle of host publication	7th International Conference on Pattern Analysis and Intelligent Systems
Editors	Chaker Abdelaziz Kerrache, Makhlouf Derdour, Nassira Ghoualmi-Zine, Bouhamed Mohammed Mounir
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798331526252
DOIs	https://doi.org/10.1109/PAIS66004.2025.11126497
State	Published - 2025
Event	7th International Conference on Pattern Analysis and Intelligent Systems, PAIS 2025 - Laghouat, Algeria Duration: 23 Apr 2025 → 24 Apr 2025

Publication series

Name	PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems

Conference

Conference	7th International Conference on Pattern Analysis and Intelligent Systems, PAIS 2025
Country/Territory	Algeria
City	Laghouat
Period	23/04/25 → 24/04/25

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

CADx
CT scans
Deep Learning
LIME
Pulmonary nodules

Access to Document

10.1109/PAIS66004.2025.11126497

Cite this

Bouamrane, A., Derdour, M., Harous, S., Abdellah, K., Deriche, M., & Kahil, M. S. (2025). Explainable Lymph Node Diagnosis Model Combining MobileNet and ConvNextTiny. In C. A. Kerrache, M. Derdour, N. Ghoualmi-Zine, & B. M. Mounir (Eds.), PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems (PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PAIS66004.2025.11126497

Bouamrane, Amira ; Derdour, Makhlouf ; Harous, Saad et al. / Explainable Lymph Node Diagnosis Model Combining MobileNet and ConvNextTiny. PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems. editor / Chaker Abdelaziz Kerrache ; Makhlouf Derdour ; Nassira Ghoualmi-Zine ; Bouhamed Mohammed Mounir. Institute of Electrical and Electronics Engineers Inc., 2025. (PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems).

@inproceedings{d3e6fd89ac2a4f6faa0ef1ec5be308eb,

title = "Explainable Lymph Node Diagnosis Model Combining MobileNet and ConvNextTiny",

abstract = "Computer-Aided Diagnostic (CADx) systems have proven effective in classifying pulmonary nodules. However, these models' reliability issue represents a topic of research and discussion. This paper proposes an interpretable lightweight hybrid model for diagnosing lymph nodes using computed tomography images and deep learning. The hybrid model combines MobileNetV3Small with ConvNextTiny to highlight features and improve performance by combining two datasets to assess data diversity, including LIDC-IDRI and the chest CT scan images for the lung cancer dataset. The model is evaluated using an external dataset to show its generalization capability. LIME is used to explain the model's decisions. The approach achieved a low false positive rate (FPR) and false negative rate (FNR) of 1.04\%, resulting in high performance in all the metrics with an accuracy of 98.56\% and a receiver operating characteristic (ROC) of 100\% in the IQ-OTH/NCCD dataset.",

keywords = "CADx, CT scans, Deep Learning, LIME, Pulmonary nodules",

author = "Amira Bouamrane and Makhlouf Derdour and Saad Harous and Kouzo Abdellah and Mohamed Deriche and Kahil, \{Moustafa Sadek\}",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 7th International Conference on Pattern Analysis and Intelligent Systems, PAIS 2025 ; Conference date: 23-04-2025 Through 24-04-2025",

year = "2025",

doi = "10.1109/PAIS66004.2025.11126497",

language = "English",

series = "PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

editor = "Kerrache, \{Chaker Abdelaziz\} and Makhlouf Derdour and Nassira Ghoualmi-Zine and Mounir, \{Bouhamed Mohammed\}",

booktitle = "PAIS 2025 - Proceeding",

address = "United States",

}

Bouamrane, A, Derdour, M, Harous, S, Abdellah, K, Deriche, M & Kahil, MS 2025, Explainable Lymph Node Diagnosis Model Combining MobileNet and ConvNextTiny. in CA Kerrache, M Derdour, N Ghoualmi-Zine & BM Mounir (eds), PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems. PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems, Institute of Electrical and Electronics Engineers Inc., 7th International Conference on Pattern Analysis and Intelligent Systems, PAIS 2025, Laghouat, Algeria, 23/04/25. https://doi.org/10.1109/PAIS66004.2025.11126497

Explainable Lymph Node Diagnosis Model Combining MobileNet and ConvNextTiny. / Bouamrane, Amira; Derdour, Makhlouf; Harous, Saad et al.
PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems. ed. / Chaker Abdelaziz Kerrache; Makhlouf Derdour; Nassira Ghoualmi-Zine; Bouhamed Mohammed Mounir. Institute of Electrical and Electronics Engineers Inc., 2025. (PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Explainable Lymph Node Diagnosis Model Combining MobileNet and ConvNextTiny

AU - Bouamrane, Amira

AU - Derdour, Makhlouf

AU - Harous, Saad

AU - Abdellah, Kouzo

AU - Deriche, Mohamed

AU - Kahil, Moustafa Sadek

PY - 2025

Y1 - 2025

N2 - Computer-Aided Diagnostic (CADx) systems have proven effective in classifying pulmonary nodules. However, these models' reliability issue represents a topic of research and discussion. This paper proposes an interpretable lightweight hybrid model for diagnosing lymph nodes using computed tomography images and deep learning. The hybrid model combines MobileNetV3Small with ConvNextTiny to highlight features and improve performance by combining two datasets to assess data diversity, including LIDC-IDRI and the chest CT scan images for the lung cancer dataset. The model is evaluated using an external dataset to show its generalization capability. LIME is used to explain the model's decisions. The approach achieved a low false positive rate (FPR) and false negative rate (FNR) of 1.04%, resulting in high performance in all the metrics with an accuracy of 98.56% and a receiver operating characteristic (ROC) of 100% in the IQ-OTH/NCCD dataset.

AB - Computer-Aided Diagnostic (CADx) systems have proven effective in classifying pulmonary nodules. However, these models' reliability issue represents a topic of research and discussion. This paper proposes an interpretable lightweight hybrid model for diagnosing lymph nodes using computed tomography images and deep learning. The hybrid model combines MobileNetV3Small with ConvNextTiny to highlight features and improve performance by combining two datasets to assess data diversity, including LIDC-IDRI and the chest CT scan images for the lung cancer dataset. The model is evaluated using an external dataset to show its generalization capability. LIME is used to explain the model's decisions. The approach achieved a low false positive rate (FPR) and false negative rate (FNR) of 1.04%, resulting in high performance in all the metrics with an accuracy of 98.56% and a receiver operating characteristic (ROC) of 100% in the IQ-OTH/NCCD dataset.

KW - CADx

KW - CT scans

KW - Deep Learning

KW - LIME

KW - Pulmonary nodules

UR - https://www.scopus.com/pages/publications/105016002931

U2 - 10.1109/PAIS66004.2025.11126497

DO - 10.1109/PAIS66004.2025.11126497

M3 - Conference contribution

AN - SCOPUS:105016002931

T3 - PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems

BT - PAIS 2025 - Proceeding

A2 - Kerrache, Chaker Abdelaziz

A2 - Derdour, Makhlouf

A2 - Ghoualmi-Zine, Nassira

A2 - Mounir, Bouhamed Mohammed

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 7th International Conference on Pattern Analysis and Intelligent Systems, PAIS 2025

Y2 - 23 April 2025 through 24 April 2025

ER -

Bouamrane A, Derdour M, Harous S, Abdellah K, Deriche M, Kahil MS. Explainable Lymph Node Diagnosis Model Combining MobileNet and ConvNextTiny. In Kerrache CA, Derdour M, Ghoualmi-Zine N, Mounir BM, editors, PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems. Institute of Electrical and Electronics Engineers Inc. 2025. (PAIS 2025 - Proceeding: 7th International Conference on Pattern Analysis and Intelligent Systems). doi: 10.1109/PAIS66004.2025.11126497

Explainable Lymph Node Diagnosis Model Combining MobileNet and ConvNextTiny

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this